[ REQ 5013]: GeomView query

[daemon]

We are using GeomView for some visualization of sets of polyhedral
models.  Typically, we have several very complex polyhedra that are
stationary and we want to "fly" an object (say one of the
polyhedra) around the workspace, with translations and rotations.
(the application is collision detection)

Now, when the collision detection program calls geomview and the 
windows pop open, we've written the program so that the user must 
type a key in order to begin the "flight".  The idea behind this 
was to allow the user to manipulate the scene so that we could 
get a better view of what was about to happen.  If the user does 
not manipulate the scene at all and simply presses a key to 
continue, the program executes correctly and we can see exactly 
which stationary polyhedra is hit by the flying object.

However, if the user does manipulate the scene (rotates it or 
translates it etc.) then we no longer see what is actually happening.
That is, the flying object appears to fly its course but not in
relation to the location of the manipulated stationary objects. 
Also, if we manipulate the scene during the flight, the same 
problem occurs.

It would appear that the flying object is no longer located in
the same position relative to the other objects.  When we manipulate
the scene we are effecting every object (the flying object and the
stationary objects) so it is not simply a matter of moving everything 
else.

Do you have any ideas as to why this would happen if (as I've already
said) we always manipulate the entire scene?

We have noticed that if we manipulate the scene before the flight
to get a better look at the objects and then click the mouse on
"Reset" in the "Tools" window, the scene is restored to its 
initial position and the flight will proceed correctly.

We assumed that as long as we manipulate everything, then the 
objects should remain in the same positions relative to one 
another and therefore manipulations (before or during the flight)
should not be a problem in this application.

Any thoughts or ideas?  Any help would be greatly appreciated.

If I've done a lousy job of explaining the problem, please
let me know and I'll try to explain it a little more clearly
and concisely.  It is not very easy to do using email.

Thank you very much.

Jim Klosowski
jklosow at mathlab.sunysb.edu